New 1,4-anthracene-9,10-dione derivatives as potential anticancer agents

Article abstract of DOI:10.1016/S0014-827X(99)00091-9

The amino-substituted anthracene-9,10-dione (9,10-anthraquinone) derivatives represent one of the most important classes of potential anticancer agents. To better understand the basic rules governing DNA sequence specificity, we have recently synthesized a new class of d- and l- aminoacyl-anthraquinone derivatives. We have tested these new compounds as cytotoxic agents, and we have correlated their activity with the configuration of the chiral aminoacyl moiety. Molecular modeling studies have been performed to compare the test drugs in terms of steric overlapping. (C) 2000 Elsevier Science S.A.

Full text of DOI:10.1016/S0014-827X(99)00091-9

www.elsevier.nl/locate/farmac
Il Farmaco 55 (2000) 1–5
New 1,4-anthracene-9,10-dione derivatives as potential anticancer
agents
G. Zagotto a,*, R. Supino b, E. Favini b, S. Moro a, M. Palumbo a
a Dipartimento di Scienze Farmaceutiche, Uni6ersita’ di Pado6a, 6ia Marzolo 5, I-35131 Padua, Italy
b Istituto Nazionale per lo Studio e la Cura dei Tumori, Di6isione di Oncologia Sperimentale B, 6ia Venezian 1, I-20133 Milan, Italy
Received 15 July 1999; accepted 10 September 1999
Abstract
The amino-substituted anthracene-9,10-dione (9,10-anthraquinone) derivatives represent one of the most important classes of
potential anticancer agents. To better understand the basic rules governing DNA sequence specificity, we have recently synthesized
a new class of D- and L-aminoacyl–anthraquinone derivatives. We have tested these new compounds as cytotoxic agents, and we
have correlated their activity with the configuration of the chiral aminoacyl moiety. Molecular modeling studies have been
performed to compare the test drugs in terms of steric overlapping. © 2000 Elsevier Science S.A. All rights reserved.
Keywords: Anthracene-9,10-dione; DNA; Specificity; Topoisomerase; Molecular modeling
1. Introduction
we have recently investigated a new class of
D- and
L
-aminoacyl–anthraquinone derivatives [7,8]. In this
The amino-functionalized anthracene-9,10-dione
(9,10-anthraquinone) derivatives represent one of the
most important classes of anticancer agents. For exam-
ple, mitoxantrone (1,4-dihydroxy-5,8-bis[[2-[(2-hydroxy-
ethyl)amino]ethyl]amino]-9,10-anthracenedione) is now
licensed for clinical use in a number of countries against
breast cancer and acute leukemias. As already exten-
sively reported, the main target of these 9,10-an-
thraquinone derivatives in the cell appears to be DNA,
with which they interact by intercalation. However, the
mode of action of anthracene-9,10-dione derivatives is
believed to involve the production of double-strand
DNA breaks mediated by DNA topoisomerase II.
DNA topoisomerases are essential enzymes that modify
the topology of nucleic acids. Agents able to interfere
with the DNA–enzyme recognition mechanisms have
been recently synthesized [1–5] and their DNA-binding
and antitumor properties evaluated. To better under-
stand the basic rules of the DNA sequence specificity [6]
paper, as a confirmation of an investigation aimed at
defining the role of aminoacyl substitution and chirality
on biological activity, we have synthesized a series of
tyrosine derivatives both in the L- and D-configuration.
2. Experimental
Melting points were determined by a Gallenkamp
apparatus, and are uncorrected. FT-IR spectra were
recorded on a Perkin–Elmer System 1760 FT-spec-
trometer using KBr mulls. 1H NMR spectra were
recorded on a Varian Gemini 200 (200 MHz) appara-
tus. Elemental analyses (CHN) were within 0.40% of
the calculated values and were performed on a Carlo
Erba 1016 Elemental Analyser. Optical purities of the
amino acid derivatives were determined by a Perkin–
Elmer 141 Polarimeter and compared, when possible,
with known values [9,10]. UV–Vis spectra were
recorded on a Perkin–Elmer Lambda 5 spectrophoto-
meter and fluorescence measurements on a Perkin–
Elmer Luminescence Spectrometer LS 50. Reagents and
solvents were used as purchased without further purifi-
cation. Silica gel and TLC plates were from E. Merck.
* Corresponding author. Tel.: +39-049-827-5686; fax: +39-049-
827-5366.
E-mail address: giuseppe@purple.dsfarm.unipd.it (G. Zagotto)
0014-827X/00/$ - see front matter © 2000 Elsevier Science S.A. All rights reserved.
PII: S0014-827X(99)00091-9

2
G. Zagotto et al. / Il Farmaco 55 (2000) 1–5
1
Melting points are in °C, IR frequencies in cm−1, H
1614, 1585, 1514, 1289, 1244, 1168, 1027. UV: 599.7,
558.9, 251.1, 249.7, 228.4, 204.9. Mass M+=373.
NMR chemical shifts (l) in ppm and coupling constants
(J) in Hz. IR wave numbers are in cm−1 and UV
frequencies in nm.
2.2. DNA binding
Measurements were carried out at 25°C in ETN buffer
(1 mM EDTA, 10 nM Tris, pH 7.0). Binding was
monitored spectrophotometrically or fluorimetrically in
the ligand absorption or emission region.
2.1. Synthesis
L
- and D-tyrosine methyl esters were obtained by
refluxing the amino acid in anhydrous methanol with
HCl as catalyst, and reducing the alcohol with sodium
borohydride in an ethanol–water solution.
2.3. Cell toxicity
2.3.1. Cell lines and in 6itro experiments
2.1.1. General procedure for the synthesis of the
anthraquinone deri6ati6es
Cytotoxic effects were evaluated on the following cell
lines: H460 human non-small cell lung carcinoma (kindly
provided by Dr Gazdar, National Cancer Institute and
Naval Hospital, Bethesda, MD, USA; PC3 human
prostate carcinoma (obtained from American Tissue
Culture Society). All cell lines were maintained in RPMI-
1640 containing 10% fetal calf serum.
To perform the experiments cells were seeded at a
density of 3000–5000 cell/well, 100 ml/well in 96-well
plates, and treated 24 h later for 24 h with the drugs. Then
the medium was washed out and cells were incubated in
drug-free medium for a further 48 h. Then cell survival
was evaluated by sulforhodamine B (SRB) assay accord-
ing to Skehan et al. [11]. Samples were fixed with 10%
(w/v) trichloroacetic acid and cellular proteins were
stained with 0.4% SRB dissolved in acetic acid. Protein-
bound dye was solubilized with 10 mM Tris, pH 10.4 and
optical density at 550 nm was measured with a microplate
reader (EAR 400 AT; SLT-Labinstruments, Austria).
IC50 was defined as the drug concentration causing a 50%
reduction of absorbance at 550 nm.
Potassium carbonate (10 g, 72 mmol) in water (150 ml)
was warmed up to 80°C and oxygen was carefully
removed from the solution by fluxing dry nitrogen. To
the deaerated solution sodium hydrosulfite (sodium
dithionite, 8.7 g, 50 mmol) and 1,4-dihydroxyan-
thraquinone (quinizarine, 7.5 g, 31 mmol) were added.
During the reaction, further sodium hydrosulfite (5 g, 29
mmol) was added in small portions. The suspension was
filtered and the solid washed and dried to give 7 g (93%)
of 2,3-dihydroquinizarine.
A solution of the tyrosyl derivative, ester or alcohol,
(2.6 mmol) dissolved in 100 ml of DMF was deaerated
and triethylamine (2.6 mmol) was dropped in to give the
free amino acid. 2,3-Dihydroquinizarine (1.3 mmol) was
added to the solution. The reaction mixture was warmed
up to 120°C for 5 h and the advancement of the reaction
was determined by TLC (DCM). When equilibrium was
reached the reaction mixture was cooled, the solvent was
removed under reduced pressure and the residue parti-
tioned in DCM and water. The organic phase was
washed three times with water, dried over sodium sulfate
and evaporated to dryness. The residue was purified by
column chromatography (silica gel, from DCM to 95:5
DCM–ethyl acetate).
2.4. Computational methods
All anthracenyl–aminoacid models were constructed
using the ‘Molecule Builder’ module of Molecular Oper-
ating Environment (MOE 1998.10) [12]. These structures
were minimized using the MMFF94 force field [13–17],
until the rms value of the Truncated Newton method
2.1.2. 2-(4-Hydroxy-9,10-dioxo-9,10-dihydro-
anthracen-1-ylamino)-3-(4-hydroxyphenyl)-propionic
acid methyl ester (L, D)
NMR (CDCl3): 13.6 (s, 1H), 10.6 (d, J=7.7, 1H), 8.4
(m, 2H), 7.8 (m, 2H), 7.1 (m, 2H), 6.8 (m, 4H), 4.9 (s,
1H), 4.5 (m, 1H), 3.8 (s, 1H), 3.2 (m, 2H). IR: 3600, 3300,
3060, 1741, 1617, 1611, 1448, 1500, 1245, 1065. UV:
583.2, 545.8, 278.6, 249.9, 227.7, 202.5; m545.8=10 692
(ethanol). Mass M+=418.
,
(TN) wasB0.001 kcal/mol/A.
The optimized geometries of all anthracenyl–amino-
acid structures were fully minimized using the RHF/
AM1//RHF/3-21G(*) ab initio level of Gaussian 98 [18].
Superimposition of the geometry-optimized
D- and
L
-TyrOMe structures was carried out using the ‘Interac-
tive Superposition’ algorithm method implemented in
Molecular Operating Environment (MOE 1998.10) [12].
2.1.3. 1-Hydroxy-4-[2-hydroxy-1-(4-hydroxybenzyl)-
ethylamino]anthracene-9,10-dione (L, D)
3. Results and discussion
NMR (CDCl3): 13.7 (s, 1H), 10.6 (d, J=7.4, 1H), 8.30
(m, 2H), 7.9 (s, 1H), 7.7 (m, 2H), 7.1 (m, 4H), 6.7 (m,
2H), 3.9 (m, 1H), 3.7 (m, 2H), 2.9 (m, 3H). IR: 3404,
Three novel compounds have been synthesized and
chemically characterized. The synthetic procedure is

G. Zagotto et al. / Il Farmaco 55 (2000) 1–5
3
shown in Scheme 1. The chemistry is quite straight-
Results on two cancer cell lines, a human prostate
forward [19] and the low solubility of the an-
thraquinone derivatives was the only difficulty that had
to be overcome throughout the entire work. The ester
reduction, as well as the reaction of the amino acid
derivative with the dihydroquinizarine, was performed
in slightly alkaline medium to avoid racemization.
Solubility problems arose when attempting to
evaluate the DNA–anthraquinone binding constants.
In fact, even at micromolar concentrations of the test
compounds, a precipitate was observed, which pre-
vented us from making a safe quantitative evaluation.
It is likely that the carboxylate arising from the ester
hydrolysis, catalyzed by esterases, represents the active
drug in this class [20,21]. In this connection, the ester
derivatives should be regarded as prodrugs and a long
exposure of the cells to the drug is necessary to achieve
the desired cytotoxicity [22]. Since two of the
compounds considered in the present work are not
esters, the cells were exposed to them for 24 h.
Furthermore, the compounds were tested against a
Topoisomerase I inhibitor (Camptothecin, CPT) and a
Topoisomerase II inhibitor (Ametantrone, AMET); this
is because analogs of the test derivatives have been
reported as being active on both enzymes [23,24].
carcinoma (PC3) and lung carcinoma (H460), are
reported in Figs. 1 and 2. In both cell lines CPT proves
to be more potent. This is in agreement with the fact
that cell lines from solid tumors are particularly
sensitive to Topo I inhibitors. Overall, compounds
III–IV are poorly active on the test systems, their IC50
largely exceeding 1 mM. Although the
and VI appear to perform slightly better than their
-analogs, the differences do not appear to be signifi-
D-enantiomers V
L
cant enough to draw safe conclusions.
Molecular modeling studies have been employed to
rationalize the above observations. Superposition
between the geometry-optimized
D- and L-TyrOMe
structures is shown in Fig. 3. Eight equivalent atoms
from the anthraquinone ring system were overlaid,
leaving the amino acid substituent in its preferred
minimum-energy conformation for each structure. The
inversion of the a-amino acid chiral center does not
seem to dramatically alter the three-dimensional
arrangement of the two structures. In fact, the internal
H-bonds between the NH of both D-/L-amino acids and
the oxygen atom of the carbonyl in position 9 of the
anthraquinone moiety freeze the position of the phenyl
Scheme 1.

4
G. Zagotto et al. / Il Farmaco 55 (2000) 1–5
Fig. 2. Cell toxicity on a lung carcinoma (H460) of the reference
compounds (upper) and new synthesized compounds (lower).
Fig. 1. Cell toxicity on a human prostate carcinoma (PC3) of the
reference compounds (upper) and new synthesized compounds
(lower).
References
[1] C. Bailly, J.B. Chaires, Sequence-specific DNA minor groove
binders. Design and synthesis of netropsin and distamycin ana-
logues, Bioconj. Chem. 9 (1998) 513–538.
[2] G. Capranico, M. Binaschi, DNA sequence selectivity of to-
poisomerases and topoisomerase poisons, Biochim. Biophys.
Acta Gene Struct. Expr. 1400 (1998) 185–194.
[3] I.Y. Choi, I.K. Chung, M.T. Muller, Eukaryotic topoisomerase
II cleavage is independent of duplex DNA conformation,
Biochim. Biophys. Acta 1264 (1995) 209–214.
[4] D.B. Jacoby, P.C. Wensink, DNA binding specificities of YPF1,
a Drosophila homolog to the DNA binding subunit of human
DNA-dependent protein kinase, Ku, J. Biol. Chem. 271 (1996)
16827–16832.
[5] M. Palumbo, Introduction to DNA sequence-specific agents, in:
M. Palumbo (Ed.), Advances in DNA Sequence Specific Agents,
vol. 3, JAI Press, Greenwich, CT, 1998, pp. 1–6.
[6] D.M. Lilley, DNA–protein structural interactions, in: D.M.
Lilley (Ed.), Frontiers in Molecular Biology, Oxford University
Press, New York, 1995.
[7] B. Gatto, G. Zagotto, C. Sissi, C. Cera, E. Uriarte, G. Palu, G.
Capranico, M. Palumbo, Peptidyl anthraquinones as potential
antineoplastic drugs: synthesis, DNA binding, redox cycling, and
biological activity, J. Med. Chem. 39 (1996) 3114–3122.
[8] G. Zagotto, G. Mitaritonna, C. Sissi, M. Palumbo, New pep-
tidyl–anthraquinones: synthesis and DNA binding, Nucleosides
Nucleotides 17 (1998) 2135–2141.
Fig. 3. Superposition between the geometry-optimized
(colored in black) and -TyrOMe (colored in gray).
D-TyrOMe
L
ring in the same direction. Hence, they do not appear
to be structurally well distinguishable. The experimental
results seem to indicate that a modest chiral discrimina-
tion may occur.
In conclusion, a longer peptide chain is therefore
needed to properly address the issue of chiral discrimi-
nation in DNA-binding anticancer agents.
[9] P. Rinaldi, M. Wilk, Use of liquid crystal induced circular
dichroism for absolute configurational assignments of beta-
amino alcohols, J. Org. Chem. 48 (1983) 2141–2146.

G. Zagotto et al. / Il Farmaco 55 (2000) 1–5
[10] Y. Asano, A. Yamada, Y. Kato, K. Yamaguchi, Y. Hibino, K. [18] M.J. Frisch, Gaussian Inc., Pittsburg PA, 1988.
5
Hirai, K. Kondo, Enantioselective synthesis of (S)-amino acids
by phenylalanine dehydrogenase from Bacillus sphaericus: use of
natural and recombinant enzymes, J. Org. Chem. 55 (1990)
5567–5571.
[19] O. Bayer, Chinone, in: J. Houben, T. Weyl (Eds.), Methoden der
Organischen Chemie, vol. VII/3C, Georg Thieme Verlag, Stutt-
gart, 1979.
[20] J. Cummings, I. Meikle, J.S. Macpherson, J.F. Smyth, Charac-
terization of the major metabolite of the novel topoisomerase I
inhibitor NU/ICRF 505, Anti Cancer Drug Des. 11 (1996)
367–382.
[21] J. Cummings, I. Meikle, J.S Macpherson, J.F. Smyth, Determi-
nation of the novel topoisomerase I inhibitor NU/ICRF 505 and
its major metabolite in plasma, tissue and tumour by high-per-
formance liquid chromatography, J. Chromatogr. B 685 (1996)
159–164.
[22] I. Meikle, J. Cummings, J.S. Macpherson, J.F. Smyth, Induction
of apoptosis in human cancer cell lines by the novel anthracenyl
amino acid topoisomerase I inhibitor NU/ICRF 505, Br. J.
Cancer 74 (1996) 374–379.
[11] P. Skehan, R. Storeng, D. Scudiero, A. Monks, J. McMahon, D.
Vistica, J.T. Warren, H. Bokesch, S. Kenney, M.R. Boyd, New
colorimetric cytotoxicity assay for anticancer-drug screening, J.
Natl. Cancer Inst. 82 (1990) 1107–1112.
[12] Molecular Operating Environment (MOE 1998.10), Chemical
Computing Group Inc., 1255 University St, Suite 1600, Mon-
treal, Quebec, Canada, H3B 3X3, 1998.
[13] T.A. Halgren, Merck molecular force field 1. Basis, form, scope,
parameterization, and performance of MMFF94, J. Comput.
Chem. 17 (1996) 490–519.
[14] T.A. Halgren, Merck molecular force field 2. MMFF94 van der
Waals and electrostatic parameters for intermolecular interac-
tions, J. Comput. Chem. 17 (1996) 520–552.
[15] T.A. Halgren, Merck molecular force field 3. Molecular ge-
ometries and vibrational frequencies for MMFF94, J. Comput.
Chem. 17 (1996) 553–586.
[16] T.A. Halgren, R.B. Nachbar, Merck molecular force field 4.
Conformational energies and geometries for MMFF94, J. Com-
put. Chem. 17 (1996) 587–615.
[23] I. Meikle, J. Cummings, J.S. Macpherson, J.F. Smyth, Develop-
ment of anthracenyl–amino acid conjugates as topoisomerase I
and II inhibitors that circumvent drug resistance, Biochem.
Pharmacol. 52 (1996) 979–990.
[24] I. Meikle, J. Cummings, J.S. Macpherson, J.F. Smyth, Iden-
tification of anthracenyl–dipeptide conjugates as novel topoiso-
merase
I and II inhibitors and their evaluation as poten-
[17] T.A. Halgren, Merck molecular force field 5. Extension of
MMFF94 using experimental data, additional computational
data, and empirical rules, J. Comput. Chem. 17 (1996) 616–641.
tial anticancer drugs, Anti Cancer Drug Des. 10 (1995) 515–
527.
.
.